ENSEMBLE METHODS FOR CLASS IMBALANCE LEARNING - Imbalanced Learning: Foundations, Algorithms, and Applications - page 74

Table 4.3 AUC Results a
abalo. balan. cmc haber. housi. mfm mfz pima vehic. wpbc avg.

0 . 824 0 . 439 0 . 705 0 . 669 0 . 825 0 . 887 0 . 855 0 . 821 0 . 859 0 . 688 0 . 757 Bagg

0 . 811 0 . 616 0 . 675 0 . 641 0 . 815 0 . 888 0 . 795 0 . 788 0 . 854 0 . 716 0 . 760 Ada

0 . 827 0 . 435 0 . 669 0 . 645 0 . 828 0 . 880 0 . 840 0 . 821 0 . 869 0 . 677 0 . 749

RF

0 . 830 0 . 617 0 . 671 0 . 646 0 . 805 0 . 916 0 . 881 0 . 789 0 . 846 0 . 694 0 . 769 Under-Ada

0 . 842 0 . 593 0 . 676 0 . 643 0 . 820 0 . 919 0 . 889 0 . 818 0 . 855 0 . 661 0 . 772

Under-RF

0 . 817 0 . 540 0 . 675 0 . 637 0 . 821 0 . 889 0 . 779 0 . 791 0 . 855 0 . 711 0 . 751 Over-Ada

Over-RF

0 . 823 0 . 458 0 . 660 0 . 641 0 . 826 0 . 881 0 . 854 0 . 819 0 . 866 0 . 670 0 . 750

0 . 831 0 . 617 0 . 680 0 . 647 0 . 816 0 . 912 0 . 862 0 . 792 0 . 858 0 . 709 0 . 772 SMOTE

0 . 850 0 . 652 0 . 696 0 . 638 0 . 811 0 . 912 0 . 903 0 . 786 0 . 856 0 . 706 0 . 781 Chan

BRF

0 . 853 0 . 558 0 . 683 0 . 677 0 . 798 0 . 901 0 . 866 0 . 809 0 . 850 0 . 646 0 . 764

0 . 812 0 . 619 0 . 675 0 . 639 0 . 815 0 . 888 0 . 801 0 . 788 0 . 853 0 . 721 0 . 761

Asym

0 . 818 0 . 599 0 . 687 0 . 646 0 . 824 0 . 897 0 . 788 0 . 790 0 . 864 0 . 720 0 . 763 SMB

0 . 847 0 . 633 0 . 704 0 . 668 0 . 825 0 . 918 0 . 904 0 . 809 0 . 859 0 . 707 0 . 787 Easy

Cascade

0 . 828 0 . 637 0 . 686 0 . 653 0 . 808 0 . 905 0 . 891 0 . 799 0 . 856 0 . 712 0 . 778

a Data is adapted from [6].